Provision of long-term care for elderly persons who have become frail and dependent is one of our most pressing public policy issues. Because nursing home care is expensive it is especially important to understand the role of the family as an alternative source of care. The majority elderly who become dependent live in the community because of the support of family, yet little is known about how families, especially the adult children of the elderly, respond to the entire sequence of decline in health and functional status experienced by many elderly, from the point at which a parent first becomes dependent through to death. This research will study children's knowledge of their parents' health and living arrangements, especially the time of nursing home entry, and thus investigate the quality of children's proxy information about their parents' health. The RAND Elderly Parents Supplement to the 1991 PSID presents a unique opportunity, because adult children report information about the family's response to the parents' changes in health and also serve as proxies in reporting the elderly parents' health, functional and economic status changes over time. Because the PSID panel has surveyed these families for two decades, (1) in many cases multiple children report information on the same parent, and (2) for another subset of parents there is also self-reported primary data from parents about their own health, functional, and socioeconomic status because the parents themselves are PSID household heads and spouses. This project will exploit this unique aspect of the data to study children's knowledge of their parents' age, education, health and living arrangement. In addition, we will explore whether the accuracy of children's knowledge is associated with covariates such as geographic proximity, education and parental financial resources.